Error: "kernel: Failed to write core file for process python2.7 (error 14)"
After applying the 9.3.1-Stable-201604140566 on 2 different machines both became inaccessible via the console, ssh and the webgui. The subject error continuously repeated on the screen. I had to roll back to the previous boot environment (9.3.1-Stable-201604120241) to recover, so I was unable to get any logs or debugging info.
#1 Updated by Sean Fagan over 5 years ago
- Status changed from Unscreened to Screened
grr. sigh. Okay, I haven't seen this, so let's try something that has worked before with something like this:
- From the boot selector, select 9.3.1-Stable-201604140566
- Boot single-user, press enter when it asks for which shell.
- Run "mount -u /; mount /boot/grub"
- Then do "beadm activate FreeNAS-9.3.1-Stable-201604120241"
- Then "shutdown -r now"
N.B. I am attempting to reproduce this now, just in case.
Meanwhile, other questions to ask are whether you have disk space, and make sure there are no errors with the pool (doing "zpool scrub freenas-boot" from the previous environment, and then "zpool status freenas-boot" to check on it).
The fact that it reproduces on two different machines troubles me greatly.
#3 Updated by John Higham over 5 years ago
Experienced same issue/error after attempting to apply update today. Do not believe update finished successfully? Machine was still displaying update in progress box on web GUI after several hours while Error: "kernel: Failed to write core file for process python2.7 (error 14)" was displayed. Rebooted machine. Machine unresponsive with same error being displayed. Rebooted machine. Reverted to older boot and now performing normally. Ran scrub and status on boot. Negative issues found.
#4 Updated by Sean Fagan over 5 years ago
Right. Let's try something.
Boot into 9.3.1-Stable-201604140566. Normal boot into it (as you've already done, John Higham). After doing this, you'll want to make it the default one, by using "beadm activate FreeNAS-9.3.1-Stable-201604140566". Then let's get rid of the new one, whih should be called FreeNAS-9.3.1-Stable-201604120241 (you can verify with "beadm list"), by doing "beadm destroy -F FreeNAS-9.3.1-Stable-201604120241".
Then we get a bit creative:
rm -rf /var/db/system/update
freenas-update -C /var/db/system/update check
freenas-update -C /var/db/system/update update --reboot
After that, assuming it completes without error, reboot the system.
I'm going to try to reproduce it again, making sure I have the right version to start with.
#6 Updated by Sean Fagan over 5 years ago
Okay! Good thing I mentioned waiting.
So, do everything I said above, except with the -R / --reboot option.
I started with an initial install of 9.3.1-Stable-201604120241, upgraded to 9.3.1-Stable-201604140566, had the reported problem. I have changed the manifest on the update, so re-downloading it again will avoid the problem.
#7 Updated by Greg DePasse over 5 years ago
sorry, I got sidetracked with some other stuff last night. I saw from jkh's forum post that there was an issue with it being "rebootless" or something. After reverting to a previous BE yesterday and applying the new update this morning, all seems to be back to normal.
If you need anymore info for this though, please let me know.